UmfPackSupport.h
1// This file is part of Eigen, a lightweight C++ template library
2// for linear algebra.
3//
4// Copyright (C) 2008-2011 Gael Guennebaud <gael.guennebaud@inria.fr>
5//
6// This Source Code Form is subject to the terms of the Mozilla
7// Public License v. 2.0. If a copy of the MPL was not distributed
8// with this file, You can obtain one at http://mozilla.org/MPL/2.0/.
9
10#ifndef EIGEN_UMFPACKSUPPORT_H
11#define EIGEN_UMFPACKSUPPORT_H
12
13namespace Eigen {
14
15/* TODO extract L, extract U, compute det, etc... */
16
17// generic double/complex<double> wrapper functions:
18
19inline void umfpack_free_numeric(void **Numeric, double)
20{ umfpack_di_free_numeric(Numeric); *Numeric = 0; }
21
22inline void umfpack_free_numeric(void **Numeric, std::complex<double>)
23{ umfpack_zi_free_numeric(Numeric); *Numeric = 0; }
24
25inline void umfpack_free_symbolic(void **Symbolic, double)
26{ umfpack_di_free_symbolic(Symbolic); *Symbolic = 0; }
27
28inline void umfpack_free_symbolic(void **Symbolic, std::complex<double>)
29{ umfpack_zi_free_symbolic(Symbolic); *Symbolic = 0; }
30
31inline int umfpack_symbolic(int n_row,int n_col,
32 const int Ap[], const int Ai[], const double Ax[], void **Symbolic,
33 const double Control [UMFPACK_CONTROL], double Info [UMFPACK_INFO])
34{
35 return umfpack_di_symbolic(n_row,n_col,Ap,Ai,Ax,Symbolic,Control,Info);
36}
37
38inline int umfpack_symbolic(int n_row,int n_col,
39 const int Ap[], const int Ai[], const std::complex<double> Ax[], void **Symbolic,
40 const double Control [UMFPACK_CONTROL], double Info [UMFPACK_INFO])
41{
42 return umfpack_zi_symbolic(n_row,n_col,Ap,Ai,&internal::real_ref(Ax[0]),0,Symbolic,Control,Info);
43}
44
45inline int umfpack_numeric( const int Ap[], const int Ai[], const double Ax[],
46 void *Symbolic, void **Numeric,
47 const double Control[UMFPACK_CONTROL],double Info [UMFPACK_INFO])
48{
49 return umfpack_di_numeric(Ap,Ai,Ax,Symbolic,Numeric,Control,Info);
50}
51
52inline int umfpack_numeric( const int Ap[], const int Ai[], const std::complex<double> Ax[],
53 void *Symbolic, void **Numeric,
54 const double Control[UMFPACK_CONTROL],double Info [UMFPACK_INFO])
55{
56 return umfpack_zi_numeric(Ap,Ai,&internal::real_ref(Ax[0]),0,Symbolic,Numeric,Control,Info);
57}
58
59inline int umfpack_solve( int sys, const int Ap[], const int Ai[], const double Ax[],
60 double X[], const double B[], void *Numeric,
61 const double Control[UMFPACK_CONTROL], double Info[UMFPACK_INFO])
62{
63 return umfpack_di_solve(sys,Ap,Ai,Ax,X,B,Numeric,Control,Info);
64}
65
66inline int umfpack_solve( int sys, const int Ap[], const int Ai[], const std::complex<double> Ax[],
67 std::complex<double> X[], const std::complex<double> B[], void *Numeric,
68 const double Control[UMFPACK_CONTROL], double Info[UMFPACK_INFO])
69{
70 return umfpack_zi_solve(sys,Ap,Ai,&internal::real_ref(Ax[0]),0,&internal::real_ref(X[0]),0,&internal::real_ref(B[0]),0,Numeric,Control,Info);
71}
72
73inline int umfpack_get_lunz(int *lnz, int *unz, int *n_row, int *n_col, int *nz_udiag, void *Numeric, double)
74{
75 return umfpack_di_get_lunz(lnz,unz,n_row,n_col,nz_udiag,Numeric);
76}
77
78inline int umfpack_get_lunz(int *lnz, int *unz, int *n_row, int *n_col, int *nz_udiag, void *Numeric, std::complex<double>)
79{
80 return umfpack_zi_get_lunz(lnz,unz,n_row,n_col,nz_udiag,Numeric);
81}
82
83inline int umfpack_get_numeric(int Lp[], int Lj[], double Lx[], int Up[], int Ui[], double Ux[],
84 int P[], int Q[], double Dx[], int *do_recip, double Rs[], void *Numeric)
85{
86 return umfpack_di_get_numeric(Lp,Lj,Lx,Up,Ui,Ux,P,Q,Dx,do_recip,Rs,Numeric);
87}
88
89inline int umfpack_get_numeric(int Lp[], int Lj[], std::complex<double> Lx[], int Up[], int Ui[], std::complex<double> Ux[],
90 int P[], int Q[], std::complex<double> Dx[], int *do_recip, double Rs[], void *Numeric)
91{
92 double& lx0_real = internal::real_ref(Lx[0]);
93 double& ux0_real = internal::real_ref(Ux[0]);
94 double& dx0_real = internal::real_ref(Dx[0]);
95 return umfpack_zi_get_numeric(Lp,Lj,Lx?&lx0_real:0,0,Up,Ui,Ux?&ux0_real:0,0,P,Q,
96 Dx?&dx0_real:0,0,do_recip,Rs,Numeric);
97}
98
99inline int umfpack_get_determinant(double *Mx, double *Ex, void *NumericHandle, double User_Info [UMFPACK_INFO])
100{
101 return umfpack_di_get_determinant(Mx,Ex,NumericHandle,User_Info);
102}
103
104inline int umfpack_get_determinant(std::complex<double> *Mx, double *Ex, void *NumericHandle, double User_Info [UMFPACK_INFO])
105{
106 double& mx_real = internal::real_ref(*Mx);
107 return umfpack_zi_get_determinant(&mx_real,0,Ex,NumericHandle,User_Info);
108}
109
123template<typename _MatrixType>
124class UmfPackLU : internal::noncopyable
125{
126 public:
127 typedef _MatrixType MatrixType;
128 typedef typename MatrixType::Scalar Scalar;
129 typedef typename MatrixType::RealScalar RealScalar;
130 typedef typename MatrixType::Index Index;
131 typedef Matrix<Scalar,Dynamic,1> Vector;
132 typedef Matrix<int, 1, MatrixType::ColsAtCompileTime> IntRowVectorType;
133 typedef Matrix<int, MatrixType::RowsAtCompileTime, 1> IntColVectorType;
134 typedef SparseMatrix<Scalar> LUMatrixType;
135 typedef SparseMatrix<Scalar,ColMajor,int> UmfpackMatrixType;
136
137 public:
138
139 UmfPackLU() { init(); }
140
141 UmfPackLU(const MatrixType& matrix)
142 {
143 init();
144 compute(matrix);
145 }
146
147 ~UmfPackLU()
148 {
149 if(m_symbolic) umfpack_free_symbolic(&m_symbolic,Scalar());
150 if(m_numeric) umfpack_free_numeric(&m_numeric,Scalar());
151 }
152
153 inline Index rows() const { return m_copyMatrix.rows(); }
154 inline Index cols() const { return m_copyMatrix.cols(); }
155
162 {
163 eigen_assert(m_isInitialized && "Decomposition is not initialized.");
164 return m_info;
165 }
166
167 inline const LUMatrixType& matrixL() const
168 {
169 if (m_extractedDataAreDirty) extractData();
170 return m_l;
171 }
172
173 inline const LUMatrixType& matrixU() const
174 {
175 if (m_extractedDataAreDirty) extractData();
176 return m_u;
177 }
178
179 inline const IntColVectorType& permutationP() const
180 {
181 if (m_extractedDataAreDirty) extractData();
182 return m_p;
183 }
184
185 inline const IntRowVectorType& permutationQ() const
186 {
187 if (m_extractedDataAreDirty) extractData();
188 return m_q;
189 }
190
195 void compute(const MatrixType& matrix)
196 {
197 analyzePattern(matrix);
198 factorize(matrix);
199 }
200
205 template<typename Rhs>
206 inline const internal::solve_retval<UmfPackLU, Rhs> solve(const MatrixBase<Rhs>& b) const
207 {
208 eigen_assert(m_isInitialized && "UmfPackLU is not initialized.");
209 eigen_assert(rows()==b.rows()
210 && "UmfPackLU::solve(): invalid number of rows of the right hand side matrix b");
211 return internal::solve_retval<UmfPackLU, Rhs>(*this, b.derived());
212 }
213
218// template<typename Rhs>
219// inline const internal::sparse_solve_retval<UmfPAckLU, Rhs> solve(const SparseMatrixBase<Rhs>& b) const
220// {
221// eigen_assert(m_isInitialized && "UmfPAckLU is not initialized.");
222// eigen_assert(rows()==b.rows()
223// && "UmfPAckLU::solve(): invalid number of rows of the right hand side matrix b");
224// return internal::sparse_solve_retval<UmfPAckLU, Rhs>(*this, b.derived());
225// }
226
233 void analyzePattern(const MatrixType& matrix)
234 {
235 if(m_symbolic)
236 umfpack_free_symbolic(&m_symbolic,Scalar());
237 if(m_numeric)
238 umfpack_free_numeric(&m_numeric,Scalar());
239
240 grapInput(matrix);
241
242 int errorCode = 0;
243 errorCode = umfpack_symbolic(matrix.rows(), matrix.cols(), m_outerIndexPtr, m_innerIndexPtr, m_valuePtr,
244 &m_symbolic, 0, 0);
245
246 m_isInitialized = true;
247 m_info = errorCode ? InvalidInput : Success;
248 m_analysisIsOk = true;
249 m_factorizationIsOk = false;
250 }
251
258 void factorize(const MatrixType& matrix)
259 {
260 eigen_assert(m_analysisIsOk && "UmfPackLU: you must first call analyzePattern()");
261 if(m_numeric)
262 umfpack_free_numeric(&m_numeric,Scalar());
263
264 grapInput(matrix);
265
266 int errorCode;
267 errorCode = umfpack_numeric(m_outerIndexPtr, m_innerIndexPtr, m_valuePtr,
268 m_symbolic, &m_numeric, 0, 0);
269
270 m_info = errorCode ? NumericalIssue : Success;
271 m_factorizationIsOk = true;
272 }
273
274 #ifndef EIGEN_PARSED_BY_DOXYGEN
276 template<typename BDerived,typename XDerived>
277 bool _solve(const MatrixBase<BDerived> &b, MatrixBase<XDerived> &x) const;
278 #endif
279
280 Scalar determinant() const;
281
282 void extractData() const;
283
284 protected:
285
286
287 void init()
288 {
289 m_info = InvalidInput;
290 m_isInitialized = false;
291 m_numeric = 0;
292 m_symbolic = 0;
293 m_outerIndexPtr = 0;
294 m_innerIndexPtr = 0;
295 m_valuePtr = 0;
296 }
297
298 void grapInput(const MatrixType& mat)
299 {
300 m_copyMatrix.resize(mat.rows(), mat.cols());
301 if( ((MatrixType::Flags&RowMajorBit)==RowMajorBit) || sizeof(typename MatrixType::Index)!=sizeof(int) || !mat.isCompressed() )
302 {
303 // non supported input -> copy
304 m_copyMatrix = mat;
305 m_outerIndexPtr = m_copyMatrix.outerIndexPtr();
306 m_innerIndexPtr = m_copyMatrix.innerIndexPtr();
307 m_valuePtr = m_copyMatrix.valuePtr();
308 }
309 else
310 {
311 m_outerIndexPtr = mat.outerIndexPtr();
312 m_innerIndexPtr = mat.innerIndexPtr();
313 m_valuePtr = mat.valuePtr();
314 }
315 }
316
317 // cached data to reduce reallocation, etc.
318 mutable LUMatrixType m_l;
319 mutable LUMatrixType m_u;
320 mutable IntColVectorType m_p;
321 mutable IntRowVectorType m_q;
322
323 UmfpackMatrixType m_copyMatrix;
324 const Scalar* m_valuePtr;
325 const int* m_outerIndexPtr;
326 const int* m_innerIndexPtr;
327 void* m_numeric;
328 void* m_symbolic;
329
330 mutable ComputationInfo m_info;
331 bool m_isInitialized;
332 int m_factorizationIsOk;
333 int m_analysisIsOk;
334 mutable bool m_extractedDataAreDirty;
335
336 private:
337 UmfPackLU(UmfPackLU& ) { }
338};
339
340
341template<typename MatrixType>
342void UmfPackLU<MatrixType>::extractData() const
343{
344 if (m_extractedDataAreDirty)
345 {
346 // get size of the data
347 int lnz, unz, rows, cols, nz_udiag;
348 umfpack_get_lunz(&lnz, &unz, &rows, &cols, &nz_udiag, m_numeric, Scalar());
349
350 // allocate data
351 m_l.resize(rows,(std::min)(rows,cols));
352 m_l.resizeNonZeros(lnz);
353
354 m_u.resize((std::min)(rows,cols),cols);
355 m_u.resizeNonZeros(unz);
356
357 m_p.resize(rows);
358 m_q.resize(cols);
359
360 // extract
361 umfpack_get_numeric(m_l.outerIndexPtr(), m_l.innerIndexPtr(), m_l.valuePtr(),
362 m_u.outerIndexPtr(), m_u.innerIndexPtr(), m_u.valuePtr(),
363 m_p.data(), m_q.data(), 0, 0, 0, m_numeric);
364
365 m_extractedDataAreDirty = false;
366 }
367}
368
369template<typename MatrixType>
370typename UmfPackLU<MatrixType>::Scalar UmfPackLU<MatrixType>::determinant() const
371{
372 Scalar det;
373 umfpack_get_determinant(&det, 0, m_numeric, 0);
374 return det;
375}
376
377template<typename MatrixType>
378template<typename BDerived,typename XDerived>
380{
381 const int rhsCols = b.cols();
382 eigen_assert((BDerived::Flags&RowMajorBit)==0 && "UmfPackLU backend does not support non col-major rhs yet");
383 eigen_assert((XDerived::Flags&RowMajorBit)==0 && "UmfPackLU backend does not support non col-major result yet");
384
385 int errorCode;
386 for (int j=0; j<rhsCols; ++j)
387 {
388 errorCode = umfpack_solve(UMFPACK_A,
389 m_outerIndexPtr, m_innerIndexPtr, m_valuePtr,
390 &x.col(j).coeffRef(0), &b.const_cast_derived().col(j).coeffRef(0), m_numeric, 0, 0);
391 if (errorCode!=0)
392 return false;
393 }
394
395 return true;
396}
397
398
399namespace internal {
400
401template<typename _MatrixType, typename Rhs>
402struct solve_retval<UmfPackLU<_MatrixType>, Rhs>
403 : solve_retval_base<UmfPackLU<_MatrixType>, Rhs>
404{
405 typedef UmfPackLU<_MatrixType> Dec;
406 EIGEN_MAKE_SOLVE_HELPERS(Dec,Rhs)
407
408 template<typename Dest> void evalTo(Dest& dst) const
409 {
410 dec()._solve(rhs(),dst);
411 }
412};
413
414template<typename _MatrixType, typename Rhs>
415struct sparse_solve_retval<UmfPackLU<_MatrixType>, Rhs>
416 : sparse_solve_retval_base<UmfPackLU<_MatrixType>, Rhs>
417{
418 typedef UmfPackLU<_MatrixType> Dec;
419 EIGEN_MAKE_SPARSE_SOLVE_HELPERS(Dec,Rhs)
420
421 template<typename Dest> void evalTo(Dest& dst) const
422 {
423 dec()._solve(rhs(),dst);
424 }
425};
426
427} // end namespace internal
428
429} // end namespace Eigen
430
431#endif // EIGEN_UMFPACKSUPPORT_H
Base class for all dense matrices, vectors, and expressions.
Definition MatrixBase.h:50
The matrix class, also used for vectors and row-vectors.
Definition Matrix.h:129
A versatible sparse matrix representation.
Definition SparseMatrix.h:87
const Index * innerIndexPtr() const
Definition SparseMatrix.h:140
const Index * outerIndexPtr() const
Definition SparseMatrix.h:149
const Scalar * valuePtr() const
Definition SparseMatrix.h:131
void resize(Index rows, Index cols)
Definition SparseMatrix.h:512
A sparse LU factorization and solver based on UmfPack.
Definition UmfPackSupport.h:125
void compute(const MatrixType &matrix)
Definition UmfPackSupport.h:195
void factorize(const MatrixType &matrix)
Definition UmfPackSupport.h:258
void analyzePattern(const MatrixType &matrix)
Definition UmfPackSupport.h:233
const internal::solve_retval< UmfPackLU, Rhs > solve(const MatrixBase< Rhs > &b) const
Definition UmfPackSupport.h:206
ComputationInfo info() const
Reports whether previous computation was successful.
Definition UmfPackSupport.h:161
ComputationInfo
Definition Constants.h:367
@ NumericalIssue
Definition Constants.h:371
@ InvalidInput
Definition Constants.h:376
@ Success
Definition Constants.h:369
const unsigned int RowMajorBit
Definition Constants.h:48
Definition LDLT.h:18